Within-Host Fine-Scale Evolutionary Dynamics of Infectious Bacterial Populations

Bacteria have a remarkable ability to adapt to numerous challenges in their environment and this evolutionary potential of bacterial pathogens has dire negative impact on human health. Understanding the fine-scale evolutionary dynamics of bacterial populations over the course of infections will help improve therapeutic strategies. In cystic fibrosis (CF) patients, lung infections caused by opportunistic pathogens such as Pseudomonas aeruginosa, Burkholderia multivorans, and Burkholderia dolosa are associated with high rates of mortality. In this thesis, I present a high-resolution view of the evolutionary dynamics of these three individual bacterial populations longitudinally sampled from CF patients. First, I analyzed the genomes of 233 P. aeruginosa isolates collected over the course of one year. This work revealed the expansion of two sub-lineages. I identified various mutations in these sub-lineages occurring in loci previously known to be associated with virulence and antibiotic resistance. Next, I investigated 111 B. multivorans isolates collected from a CF patient over the course of an early incident, chronic infection spanning one year, and after lung-transplantation. I employed a genome wide association study approach to find genetic variants associated with resistance to three antibiotic families. I observed that variants associated with resistance to β-lactam antibiotics were overrepresented in recombinogenic regions. Lastly, I examined whether the fine-scale evolutionary dynamics of B. dolosa, the most abundant bacteria in the lungs of the studied patient, was associated with shifts in the host’s clinical status from clinical stability to pulmonary exacerbation. Preliminary results do not suggest a correlation between the evolution of B. dolosa with the onset of a pulmonary exacerbation. Throughout these studies, I found unprecedented depth of genotypic diversity as these CF pathogens evolve in short time scales and strong signals of parallel evolution. Overall, this thesis reveals the power of deep, longitudinal sampling and the importance of using high-resolution approaches to study within-host evolution of bacterial pathogens as they undergo rapid, extensive changes that lower-resolution studies would not capture. This work provides an in-depth view of patient specific bacterial population dynamics, which is a step towards developing therapeutic strategies in which treatment may be tailored specifically to each CF patient.Ph.D